Archive for the ‘Research methods and approaches’ Category

Protest against Lab Testing at UXLX in Lisbon

Friday, May 17th, 2013

Protest Signs

Today we protested outside the UXLX conference to Stop Lab Testing (Usability Lab Testing).

Why protest?
I co-founded Webnographer with James, because I believed that we needed new techniques and methods to carry out user research.

The lab testing method is over 20 years old now.  Since then there has been a large shift in the technology we use, and our behaviour. We have moved on from using a desktop computer in one fixed environment 20 years ago, to portable devices (smart phones, tablet PCs, and laptops) in a multitude of contexts today. This has changed how and where people access the internet, and how much distraction or attention is given to an interaction in a given environment.

With this huge shift in tools and behaviour, the methods we use to understand individual behaviours needs to change too.

We need methods that help evaluate behaviour in its multitude of contexts, environments, languages, and countries. We need to test products with a multitude of customers, not just 10 people in London, or Lisbon, or Berlin. We need to get feedback independent of where people are located. We need to test with large numbers of users in diverse locations to be able to quantify the impact of design changes.

As Marshall McLuhan said: “We shape our tools and thereafter our tools shape us.”

This means that User Experience Research must be reshaped too. At Webnographer we are building those tools to help you understand people, so that you can make better products too.

What usability experts can learn from cockroaches

Thursday, April 21st, 2011

James recently gave a talk at the Polish IA Summit on How Context Effects Behaviour. One of the stories from his presentation outlines that there is a lot that usability experts can learn from cockroaches. The background of the story comes from Dan Ariely’s latest book “The Upside of Irrationality”.

Usability experts often forget the affect context has on human behaviour. Most usability evaluations are still carried out in labs with an evaluator peering over the participants shoulder. And the affect that the evaluator may have on the participant’s performance is often forgotten about.

But even simple creatures such as cockroaches are affected by another cockroach watching them while carrying out a task.

In his book, Dan describes an experiment that was carried out by 3 scientists in 1969. Zajonc, Heingartner, and Herman evaluated how fast cockroaches could accomplish a task in 2 different test conditions. In the first, they were alone. No other cockroaches were around. In the second, the social condition, they had an audience. Another cockroach was watching them through a Plexiglas window that allowed both creature to see and smell one another, but that did not allow any direct contact.

What the experiment found was that, carrying out an easy task – simply walking down the length of a corridor – the performance of the cockroach in the social condition increased. They walked down the corridor faster.

However, when carrying out a difficult task – navigating through a complex maze – the cockroach in the social condition performed worse.

This is an example how context affects behaviour. It is not just a trait that affects humans, but even for cockroaches’ behaviour is affected by changing external conditions.

Of course Dan Ariely does not generalise the findings from research with cockroaches to humans. He carried out many other experiments that looked at how bonuses affect behaviour. Bonuses change the social context too. People plan in advance what they will do with their bonus – go on holiday, buy a car, a present for their partner.

What Dan found in his experiments was that bonuses are a tricky business to get right. If the bonus is too great, the pressure on the person becomes too great and they actually perform worse.

Changing the social context in which we evaluate websites and design, will change the way people behave.

Lab testing has value to get in depth interviews with participants. However, for evaluating people’s real behaviour on a website as well as their performance, usability experts should look to compliment their tool box with methods such as our un-moderated remote usability testing tool Webnographer, where people participate from their natural environment without an evaluator being present.

References:

Dan Ariely “The upside of irrationality – The unexpected Benefits of Defying Logic at Work and at Home,” (2010): 17-52

Robert Zajonc, Alexander Heingartner, and Edward Herman, “Social Enhanacement and Impairement of performance in the Cockraoch,” Journal of Personality and Social Psychology 13, no. 2 (1969): 83-92

Related posts:

Why exams mean nothing out of context

Why exams mean nothing out of context

Tuesday, November 23rd, 2010

One of the advantages of remote usability testing is that you are using participants in their real context.

Before we started Webnographer, James and I attended a lecture given by Robert Sternberg where he told the story about the maths ability of Brazilian street kids living in the in the favelas of Recife. This story helped both of us realise the importance of carrying out usability tests in context.

Three researchers (see: Carraher, Carraher, and Schliemann 1985) carried out research with children aged 9 to 15.  These kids had dropped out of school, and were selling sun screen, and chewing gum on the streets. The researchers worked out that they could set the kids questions by purchasing goods off them. For example, 1,000 minus 300  is the same as giving the kid a 1,000 Cruzeiros note for a product that costs 300 Cruzeiros. Multiplication can be done by asking the kids how much 3 of a product would cost.

In these tests the Brazilian street kids scored 98%. But when they were put into a formalised test setting, and asked instead of how much would 3 apples cost or what 3×9 is, the kids performance dropped to just 37%.

What is scary is that the researchers later tested middle class children in a private school. These kids did very well in the formal exam. But when they had to do transactions with real money in the street, using the same maths, they failed in being able to do the transactions.

This is one example of context and how it affects behaviour. We see it again and again with remote usability tests, that the context of the user really affects their performance. Test people where they normally use a website, and test people in an artificial environment like a lab, and behaviour will be very different.

“How to plan your remote usability test” Poster

Tuesday, October 26th, 2010

Sometimes is hard to show how to structure all the steps needed in a method so that it is easy and simple to understand. This poster, which you can download here (PDF), gives an overview of the process of conducting a remote usability testing project.

The process starts with the Research Question. It is important to step back and first think about what one wants to find out, before defining tasks for the study.

Once we know what we want to find out, we have to choose which research method to apply. There are many remote research methods that can be used, and you should be choosing the method based on what you want to find out.

There are synchronous methods, where the evaluator and participant are in different locations, but communicate during the research with one another.  And there are asynchronous methods where the evaluator and participant are separated in time and space. This includes surveys, card sorting, web analytics, diary studies and remote ethnography. Webnographer falls into the category of asynchronous methods. It allows you to assess users in their natural environment (on their laptops, at work, at home, with friends), without moderator intervention.

One of the most important issues, that we have to deal with when using data from real people, is the Legal assessment of the data collection. For example, if you are testing in Europe it is important to follow the European Data Protection Directives, if you are testing with children you will have to think about how to get parental consent. Having said that the legal site is usually very easy. A quick check for each country on which your research focuses should do the trick. However, no matter how liberal the local data collection rules are, getting informed consent from people is best practice when carrying out any kind of research.

As soon as all that is sorted out we have to think on how we are going to get people to participate on the study. Depending on the type of study, the Participant Recruitment can be carried out from the client’s own site or an advert can be posted elsewhere targeting the profiles of users that we want to get to participate.

Getting the right people to participate is important. Don’t just get any one! You need to get the right people. You wouldn’t want to be testing a website aimed at youths with a bunch of pensioners in a care home. This is an extreme example, but highlights the importance of getting the recruitment right.

By now we have chosen the tool and the people, so we have to put down our Study Design so we can structure the way we want to ask and how we want to get the answers for our research question.

A good study design isn’t finished until is tested and nothing better than run a Pilot Test to get a sense how well is our study design getting the answers to our questions.

If everything is OK than we get a green light to go live with the research and start the Data Collection that will provide us with users answers and feedback. If you have chosen to carry out synchronous remote research, this is where the hard part begins, moderating the research with each participant. For any asynchronous study, this step only entails the monitoring of the data collection.

The data collected data is then structured for analysis, so that we can Analyse the user behavior for our study. Depending on the method you chose, the analysis will be carried out in different ways to provide Insight into your initial research question.

If you would like a copy of the poster showing the remote usability testing process, you can download your copy here (PDF). Or contact us with your address and we will send you the poster via mail.

One day workshop at the UX Lisbon conference 12th to the 13th of May 2010

Friday, February 19th, 2010

Monument to Henry the Navigator

Webnographer and UX Lisbon are organising a one day workshop, spread over two days on Remote Usability between the 12th to the 13th of May. The workshop is being held in conjunction with the conference.

The purpose of both workshops is to enable you to turn data and observations from Remote Testing into valuable insights which can be actioned. After attending both workshops you will know what questions to ask the data, and be able to prioritize the feedback.

The cost of the remote workshops if booked by the 28th of February is € 595.00, the price includes entry to the Ux Lisbon Conference as well. For just € 795 you can attend an extra 2 workshops and the conference. These prices are valid until 28th of February. There is a discount of 10% for Usability Professionals Association, Information Architecture Institute, Interaction Design Association, Association for Computing Machinery, and 15% for UxBrighton members.

As we at Webnographer want to spread the message about Remote Usability, we will offer you a voucher worth €225 off your next remote usability study, valid until the end of 2010 if you attend the workshop.

To book your place for the workshop simply go to http://www.ux-lx.com/registration.html and email lisbon@webnographer.com so that your place in the workshop is confirmed.

Workshop 1: An Introduction to Remote Usability

Remote Usability is a collection of methods that over the last year has become increasingly popular. Most of us are designing and developing products that will be used anywhere in the world, but most of our research methods are tied to a physical location. Remote sets the researcher free from being tied to a place, and also enables the testing of more people. By the end of the session you will become familiar with a number of remote research technologies, from remote ethnography to synchronous and asynchronous tools, and will be able to explain when such tools are suitable for user research.

The workshop will cover :-

  • The different methods 1.30
  • Remote ethnography
  • Surveys
  • Synchronous
  • Asynchronous
  • Study design
  • Recruitment Methods
  • Basic Analysis

Workshop 2: Asynchronous Remote Usability testing in detail

Asynchronous Remote testing is when the participant is separated from the evaluator by time and place. Asynchronous testing is significantly different from the usual user research methods, in this half a day workshop James and Sabrina will take you through the steps and theory of how to carry out a study from the design to the analysis of the results. The workshop will demystify the basic statistics that will help you understand the results. You will also learn the basics about how to carry out a International Study, from the legal issues to how to deal with translation.

  • Study design
  • True intent studies
  • Task based studies
  • Simple Statistics
  • Advanced Analysis
  • Analysing task based and true intend studies
  • Importance of filtering
  • Coding responses
  • International studies
  • Legal